The invention relates to the genetic basis of diabetics.
Diabetes mellitus is among the most common of all metabolic disorders, affecting up to 11% of the population by age 70. Type I (insulin dependent diabetes mellitus or IDDM) diabetes represents about 5 to 10% of this group and is the result of a progressive autoimmune destruction of the pancreatic .beta.-cells with subsequent insulin deficiency. Type II (non-insulin dependent diabetes mellitus or NIDDM) diabetes represents 90-95% of the affected population but is much less well understood from the point of view of primary pathogenesis. Type II diabetic patients exhibit elements of both insulin resistance and relative insulin deficiency.
Alterations in glucose homeostasis are the sine qua non of diabetes mellitus and occur in both the Type I and Type II forms of the disease. In the mildest forms of diabetes this alteration is detected only after challenge with a carbohydrate load, while in moderate to severe forms of disease hyperglycemia is present in both the fasting and postprandial states. The most important tissue involved in disposal of a glucose load following oral ingestion, i.e., in the absorptive state, is skeletal muscle. (Klip 1990 Diabetes Care 13:228-243; Caro et al. 1989 Diab. Metab. Rev. 5:665-689; Bogardus 1989 Diab. Metab. Rev. 5:527-528; Beck-Nielsen 1989 Diab. Metab. Rev. 5:487-493) Skeletal muscle comprises 40% of the body mass, but has been estimated to account for between 80 and 95% of glucose disposal at high insulin concentration or following an oral glucose load. (Beck-Nielsen 1989; Baron et al. 1988 Am. J. Physiol. 255:E769-74) In insulin-treated animals, about 25% of an intravenous glucose load enters muscle within 1 minute. (Daniel et al. 1975 J. Physiol. (Lond) 247:273-288)
Skeletal muscle takes up glucose by facilitated diffusion in both an insulin-independent and insulin-dependent manner and has been shown to express relatively high levels of GLUT4 (the "insulin responsive" glucose transporter) and low levels of GLUT1 and GLUT3 (the transporters believed to be involved in basal glucose transport). (Mueckler 1990 Diabetes 39:6-11; Bell et al. 1990 Diabetes Care 13:198-208) Once inside the muscle, glucose is rapidly phosphorylated by hexokinase to form glucose 6-phosphate. Although the rate-limiting step for glucose uptake is at the level of transport, there is increasing evidence that the major control of carbohydrate metabolism is exerted after the glucose-6-phosphate step. (Mandarino 1989 Diab. Metab. Rev. 5:474-486; Felbert et al. 1977 Diabetes 26:693-699) Depending on the hormonal milieu and metabolic state, the glucose 6-phosphate can enter either anabolic or catabolic pathways. The major anabolic pathway involves conversion of the glucose to glycogen. The rate-limiting enzyme of this reaction is glycogen synthase. The activity of glycogen synthase is regulated primarily by phosphorylation and dephosphorylation and the presence of the allosteric regulator glucose-6-phosphate, although the level of expression of the enzyme must also play a role. In catabolic states, glucose is metabolized through the glycolytic pathway to pyruvate which in turn is either converted to lactate (under anaerobic conditions) or is oxidized by CO.sub.2 and acetyl-CoA. The latter reaction is catalyzed by the multienzyme complex pyruvate dehydrogenase (PDH). PDH activity is also regulated by the level of the enzyme, phosphorylation and dephosphorylation, and a number of allosteric modifiers. Most of the enzymes and proteins involved in glucose metabolism have been identified and purified, and over the past several years, several of these have been cloned at a molecular level.
The dominant hormone regulating glucose metabolism in muscle is insulin. Insulin exerts its actions through insulin, and to a lesser extent IGF-1, receptors, both of which are expressed in skeletal muscle. (Beguinot et al. 1989 Endocrinology 125:1599-1605; Sinha et al. 1987 J. Clin. Invest. 79:1330-1337; Obermaier-Kusser et al. 1989 J. Biol. Chem. 264:9497-9504; Arner et al. 1987 Diabetologia 30:437-440) Like insulin and IGF-1 receptors on other tissues, these receptors are protein tyrosine kinases which are stimulated upon insulin and IGF-1 binding. (White et al. J. Clin. Invest. 82:1151-1156) This initial insulin signal then acts through a cascade of events involving phosphorylation and dephosphorylation, as well as possible mediator generation to promote glucose uptake, stimulate metabolism and conversion of glucose to glycogen by activating glycogen synthase, and regulate a variety of intracellular enzymes involved in carbohydrate. (Yki-Jarvinen et al. 1987 J. Clin. Invest. 80:95-100; Mandarino et al. 1987 J. Clin. Invest. 80:655-663) In addition, insulin also acts at the level of muscle to modify lipid and protein metabolism through effects on membrane transport, enzyme activity and gene expression. (Kimball et al. 1988 Diab. Metab. Rev. 4:773-787; Alexander et al. 1988 Proc. Natl. Acad. Sci. USA 85:5092-5096)
In both Type I and Type II diabetes there are major alterations in the ability of peripheral tissues to take up and metabolize glucose. (DeFronzo 1988 Diabetes 37:667-687; Olefsky et al. 1988 Am. J. Med. 85:86-105; Reaven 1988 Diabetes 37:1595-1607; Nankervis et al. 1984 Diabetologia 27:497-503; Yki-Jarvinen et al. 1986 N. Engl. J. Med. 315:224-230; Hother-Nielsen et al. 1987 Diabetologia 30:834-840) These alterations affect liver, fat and muscle, as well as other tissues. In Type I diabetes, the alterations in glucose metabolism are largely secondary to insulin deficiency which has both acute and chronic effects with regard to regulation of glucose uptake and intracellular disposition metabolism. (Nankervis 1984; Yki-Jarvinen 1986; Hother-Nielsen 1987) The exact basis for impaired metabolism in muscle of Type II diabetics is less clear, but probably involves a combination of factors including a significant level of insulin resistance (due to acquired or genetic factors), as well as some component of relative insulin deficiency.
In obesity and diabetes, there are a variety of alterations in the muscle glucose homeostasis. In obese individuals without diabetes the major alteration is in oxidative glucose metabolism. (Beck-Nielsen 1989; DeFronzo 1988; Olefsky 1988) There are reduced insulin-stimulated nonoxidative glucose metabolism, a reduction in both basal and insulin-stimulated glucose oxidation and a higher rate of lipid oxidation than in lean controls. (Felber et al. 1987 Diabetologia 26:1341-1350) Glycogen synthase activity is decreased in obese individuals and may contribute to the reduced nonoxidative glucose disposal. (Bogardus et al. 1984 J. Clin. Invest. 73:1185-1190; Freymond et al. 1988 J. Clin. Invest. 82:1503-1509) In obese diabetics both oxidative and nonoxidative pathways are altered, and the latter may play a more quantitatively important role. (Beck-Nielsen 1989) In both obesity and Type II diabetes there is also decreased insulin stimulated receptor phosphorylation (Caro 1987; Obermaier-Kusser 1989; Arner 1987) decreased insulin stimulated glucose transport (Caro 1987; Felber 1987) decreased insulin-stimulated pyruvate dehydrogenase activity (Mandarino 1989), and defective insulin-stimulated glycogen synthase. (Mandarino 1989; Freymond 1988; Thorburn et al. 1990 J. Clin. Invest. 85:522-529) Untreated Type I diabetic humans and rodents show many of the same changes. (Nankervis 1984; Yki-Jarvinen 1986; Hother-Nielsen 1987; Wallberg 1989 Med. Sci. Sports. Exerc. 21:356-361) In the latter group, these tend to reverse with proper therapy, although in most studies some reduction in insulin-stimulated glucose oxidation and insulin-stimulated PDH activity persist despite therapy. Factors mediating these alterations in glucose homeostasis in diabetes and obesity are multiple and include the altered hormonal milieu, altered substrate levels, and possibly even circulating insulin antagonists. (DeFronzo 1988; Sugden et al. 1990 J. Endocrinol. 127:187-190; Leighton et al. 1990 Trends Biochem. Sci. 15:295-299) A summary of the defects in glucose metabolism in muscle in Types I and II diabetes and obesity is given in reference 4.
Although controversy exists as to the primary defect in Type II diabetes several studies suggest that the earliest detectable abnormality may be in glucose disposal by muscle. (Bogardus 1989; DeFronzo 1988; Reaven 1988; Lilloja et al. 1988 Acta Med. Scand. Suppl. 723:103-119; Ho et al. 1990 Diabetic Med. 7:31-34; Eriksson et al. 1989 N. Engl. J. Med. 321:337-343; Bogardus et al. 1989 Diabetes 38:1423-1432; Lilloja et al. 1987 Diabetes 36:1329-1335; Knowler et al. 1990 Diab. Metab. Rev. 6:1-27; Warram et al. 1990 Ann. Int. Med. 113:909-915; Martin et al. Submitted for publication) In a study initiated by Dr. J. Stuart Soeldner over 20 years ago, over 200 offspring of two Type II diabetic parents were identified and evaluated for abnormalities in glucose tolerance, glucose disposal and insulin secretion (Warram 1990; Martin) 155 of the offspring were normoglycemic at the onset of study. The individuals were subsequently followed for an average of 14 years during which time 25 developed Type II diabetes. Analysis of data gathered at entry to the study provided a unique insight as to the earliest defects detectable in prediabetic individuals. This study revealed that there were no differences in either first or second phase insulin secretory capacity in these offspring which predicted the development of diabetes. However, overall glucose disposal rate was reduced. Based on the Bergman model of glucose disposal (Bergman 1989 Diabetes 38:1512-1527), this decrease in glucose disposal rate was due to reduced insulin sensitivity (S.sub.1) and insulin independent glucose disposal (S.sub.G). Low values of S.sub.1 and/or S.sub.G were highly predictive of the subsequent development of diabetes. (Warram 1990; Martin) If one compares the normoglycemic offspring in this study with the lowest quintile of insulin sensitivity to those offspring in the highest quintile of insulin sensitivity, there was a 62-fold increase in relative risk of developing Type II diabetes during follow-up (see preprint of Martin in Appendix). Likewise low glucose sensitivity (which reflects both insulin-independent and basal insulin-dependent glucose disposal) increased the relative risk of diabetes by 22-fold.
Similar findings and conclusions have been derived from studies of the Pima Indian population which has a very high incidence of Type II diabetes. (Lilloja 1988; Ho 1990; Erikkson 1989; Bogardus 1989; Lilloja 1987; Knowler 1990) In addition, in this population, evidence has been presented that insulin sensitivity is inherited in family clusters and with a pattern of distribution suggestive of an autosomal dominant trait. (Bogardus 1989; Lilloja 1987) Alterations in glucose metabolism and glycogen synthase activity can also be detected in biopsies from prediabetic Pima Indians prior to onset of clinical diabetes. (Knowler 1990; Warram 1990; Martin; Bergman 1989; Foley 1988 Diabet. Metabl. Rev. 4:487-505) Consistent with the hypothesis that Type II diabetes develops in the presence of an underlying defect in insulin sensitivity are the many observations which indicate that aggressive insulin therapy in the Type II diabetic may normalize blood glucose and glycohemoglobin, but usually fails to reverse completely the insulin resistance which is observed in this disease. (DeFronzo 1988; Olefsky 1988) Genetic variability in insulin sensitivity may also exist in the non-diabetic Caucasian populations (Hollenbeck et al. 1987 J. Clin. Endocrinol. Metab. 64:1169-1173), suggesting that genes which control insulin sensitivity may exist at some frequency in control non-diabetic, as well as diabetic, population.
Although there is considerable evidence for alterations in the metabolism and the activity of a variety of enzymes and transporters in muscle in diabetes mellitus, including changes in insulin receptors, glucose transporters, glycogen synthase and pyruvate dehydrogenase, there is less information concerning specific alterations in the expression of the genes for these proteins or other proteins which might account for the altered insulin action. Alterations in the level of mRNA for the insulin-sensitive glucose transporter GLUT4, but not GLUT1, have been observed in skeletal muscle of streptozotocin diabetic rats and are restored toward normal by insulin or vanadate treatment. (Bourey 1990 J. Clin. Invest. 86:542-547; Strout et al. 1990 J. Endocrinology 126:2728-2732; Sivitz 1989 Nature 340:72-74). Insulin and glucose have also been shown to regulate glucose transporter mRNA expression in the cultured skeletal muscle cell line L6. (Walker et al. 1989 J. Biol. Chem. 264:6587-6595) In both rodent and human models of obesity and Type II diabetes, the data on glucose transporter mRNA expression are more controversial with some studies showing altered levels of GLUT4 expression in adipose tissue, but not muscle, and others showing alterations in both tissues. (Caro 1989; Pedersen 1990 Diabetes 39:865-870) Diabetes has also been shown to affect the level of expression of insulin-like growth factors I and II in muscle (Leaman 1990 Endocrinology 126:2850-2857), creatine kinase (Popovich 1989 Amer. J. Physiol 257:E573-E577), glutamine synthetase (Feng 1990 Am. J. Physiol. 258:E762-E766), and the .alpha.-subunit of guaninnucleotide regulatory protein, G.sub.S. (Griffiths et al. 1990 Eur. J. Biochem. 193:357-364) By contrast, mRNA for the important bifunctional enzyme 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase is altered in liver of STZ diabetic rats, but is not altered in muscle. (Colosia 1988 J. Biol. Chem. 263:18669-18677) Some investigators have also shown a change in expression of one of the alternatively spliced forms of the insulin receptor in muscle of human with Type II diabetes, although this apparently occurs without a significant change in level of total receptor mRNA. (Mosthaf et al. 1991 Proc. Natl. Acad. Sci. USA 88:4728-4730) Levels of total mRNA and those of most of the major structural proteins like actin do not appear to be altered in diabetes. (Pedersen 1990) Little or no data yet exist for effects of diabetes on the mRNA levels in muscle for glycogen synthase, PDH, hexokinase, the insulin receptor substrate (IRS-1) etc., although gene probes now exist for a number of these important metabolic enzymes. (Persons et al. 1989 Mol. Carcinog. 2:88-94; Dugail 1988 Biochem. J. 254:483-487; Arora et al. 1990 J. Biol. Chem. 65:6481-6488; Minchenko et al. 1984 Endocrinol. Exp. 18:3-18; Sun et al. 1991 Nature 352:73-77)